Skip to contents

Build a clade nuclear-rDNA reference database (GetOrganelle seeds + labels)

Source: R/build_reference_db.R
build_reference_db.Rd

Downloads 18S / 5.8S / 28S / ITS records for a GenBank taxonomic query and builds the GetOrganelle seed + label FASTAs used by new_project() (custom_seeds_db / custom_labels_db). Run this once per clade, before running the pipeline; point every project for that clade at the result.

Usage

build_reference_db(
  query,
  clade,
  out_dir = ribopilot_ref_dir(),
  api_key = NULL,
  refseq = FALSE,
  count_only = FALSE,
  conda_env = NULL,
  overwrite = FALSE
)

Arguments

query

Advanced GenBank search term, e.g. '"Gastropoda"[Organism]' or '"Percidae"[Organism]'.

clade

Short name for the output files, e.g. "gastropoda". Files are written as <out_dir>/seeds/<clade>_rrna_seeds.fasta and <out_dir>/labels/<clade>_rrna_labels.fasta.

out_dir

Directory to store databases (default ribopilot_ref_dir(), outside any project so it is reused across projects).

api_key

Optional NCBI API key (raises the request-rate limit).

refseq

Restrict to RefSeq records.

count_only

Only print the matching-record count and return (no download).

conda_env

Optional conda env (name or path) providing Entrez Direct + biopython; its bin/ is prepended to PATH for the run.

overwrite

Rebuild even if the seed/label FASTAs already exist. When FALSE (default) an existing DB is reused (no download) and its paths returned.

Value

(invisibly) a list with seeds and labels file paths (both NA for count_only).

Details

Requires Entrez Direct + biopython (in the RiboPilot conda env). Pass conda_env to put them on PATH.

This is the CLI entry point for building a database outside the app. The RiboPilot app's Reference step builds databases through its own Nextflow workflow instead and does not call this function; use it directly for scripted or headless use, or to build a database ahead of time.